Steady, Flexible Memristor with Lead-Free Perovskite for Bionic Perception

Area of Science:

• Materials Science
• Neuroscience
• Electrical Engineering

Background:

• Artificial intelligence and cloud computing require advanced components for complex problem-solving and bionic perception.
• Perovskite memristors are emerging neural synapse devices with potential for information storage and brain-like learning due to low energy consumption and fast processing.

Purpose of the Study:

• To fabricate a flexible, lead-free perovskite memristor using low-temperature processing.
• To investigate its resistive switching characteristics, environmental stability, and mechanical performance.
• To demonstrate its capability for synaptic plasticity, fear emulation, and sound perception simulation for multimodal intelligent perception.

Main Methods:

• Fabrication of a PET/Graphene/Cs2AgBiBr6/Ag flexible memristor via low-temperature processing (~100 °C).
• Characterization of resistive switching, ON/OFF ratio (~10^2), long-term stability (>=8 months), and mechanical bending (600 cycles).
• Verification of synaptic plasticity (paired pulse facilitation, spike-timing-dependent plasticity) and emulation of fear and sound perception using voltage parameters and light illuminance.

Main Results:

• The flexible memristor exhibited pronounced resistive switching characteristics and excellent mechanical bending performance.
• Synaptic plasticity, including PPF and STDP, was successfully verified.
• The device achieved fear emulation (self-extinction, generalization, avoidance) and simulated sound perception (pitch, loudness, timbre).

Conclusions:

• The developed lead-free flexible perovskite memristor shows promise for multimodal intelligent perception and biological simulation.
• The device's characteristics are suitable for applications in artificial intelligence and brain-like learning systems.
• This work opens new application avenues for perovskite memristors in advanced computing and sensory simulation.